A Quantitative Analysis of Multimodal Biomarkers in Alzheimer’s Disease
Antonio Scardace and Daniele Ravì

This repository contains the official code for our paper on quantitative analysis of multimodal biomarkers in Alzheimer’s Disease (AD), where we systematically study how molecular, structural, clinical, and genetic biomarkers interact. Modern AD research increasingly relies on data from multiple biological domains, yet how these modalities overlap, what information each uniquely contributes, and which are most useful for prediction remain poorly understood—despite the high cost and burden these measurements place on patients.

In our paper, we analyze a cohort combining tau-PET, sMRI, MMSE, CDR, and APOE $\epsilon4$ genotype, and provides a reproducible pipeline that can be reused on similar multimodal datasets. The pipeline quantifies cross‑modal information sharing, evaluates predictive relationships between modalities, characterizes structure–pathology associations, and models the main disease progression trajectory using SuStAIn.

Citation

Please, reference this publication if you find this code useful:

    [wip]